I joined Google Brain in 2018 as a Machine Learning Engineer. My interests are in the area of machine learning, edge computing, mobile systems, and video analytics. While at Google, I have been working on the following projects :
- Quantization-aware training and transfer learning of object detection models on TPUs
- Real-time fast efficient inference of object detection models using Tensorflow Lite on mobile devices and other small platforms
- Android Demo App and Raspberry Pi demo for the detection models
This was recently released in a blog post.
I received my PhD at from Stanford University in 2013 in signal processing and communications/networking. I was a postdoctoral researcher at Microsoft Research and Princeton University between 2013-2017. My research focuses on the design on designing efficient models and compression/resource-allocation strategies to enable machine learning inference on real-time video between the IoT/mobile devices and the cloud. My research work has contributed to industry standards and consortia, such as OpenFog Consortium.
In 2012, I received the Paul Baran Marconi Young Scholar Award, given for the scientific contributions in the field of communications and the Internet. I also received the Stanford School of Engineering Fellowship and the Stanford's Diversifying Academia Recruiting Excellence (DARE) fellowship in 2010.